The hypotheses tested in these in vitro studies were: a) continuous wave Doppler (CWD) provides accurate assessment of pressure differences across prosthetic valves as a function of flow and b) color-encoded, two-dimensional Doppler aids in the interpretation of traditional flow visualization. A physiologic pulse duplicator with pressure and electromagnetic transducers, an IBM AT microprocessor, and ALOKA-SSD-880 ultrasonic system were used. Twelve mechanical aortic prostheses (1 ball valve, 2 bileaflets, and 3 tilting disc, sizes 23 & 27 mm) were studied. Flow visualization studies used a 35 mm camera, laser light source with a planoconvex lens 90 degrees to axial flow, and amberlite particles (420 M) in normal saline solution. Peak pressure differences were measured at 1 cm increments from 1 to 6 cm distal from the valve sewing ring. Simultaneous maximal velocities were obtained with CWD and transformed to pressure differences with the simplified Bernoulli equation. Manometric peak pressure differences decreased from the 1 cm (range P = 3.0 -51.2 mm Hg) to the 6 cm (range P = 1.2 - 33.0 mm Hg) loci. Maximal velocities (range 4.88 m/s to 1.21 m/s) varied by valve but correspond with maximal manometric data. Linear regression between the Doppler derived and measured pressure differences showed slopes for each valve ranging from 0.71 to 1.62, and an intercept ranging from -14.5 mm Hg to 10.5 mm Hg; 0.95 less than r less than 0.99. Flow visualization and color flow imaging revealed similar flow patterns for valves with similar geometrical configurations. It is concluded that pressure differences across prosthetic valves are a function of the location of measurement and that CWD is an accurate method of determining maximal pressure differences at known flow and cycle rates. Color flow imaging provides new accurate information about flow patterns. Ultrasonic technology provides new dynamic information about the in vitro performance characteristics of prosthetic valves.